Discrete, Class A Lateral MosFET Headphone amp (not another follower buffer!)

mnturner · 22nd February 2012, 01:27 PM

Hey guys,
I've been wanting to build a headamp for my AKG701's, which need a bit more power, for a while now but haven't seen any schems that really caught my eye.
I know the currently popular opamp based designs (eg the Wire and O2) perform exceptionally well, but I find them a little boring, and browsing Rod Elliot's site and coming across his Project 101 mosfet output power amp inspired me to design a discrete, class A mosfet headamp.
The front end is based off Self's poweramp designs sans driver stage, and the output stage consists of EXC10N20 and EXC10P20 Exicon audio Lateral Mosfets. I used these because I've had a couple of pairs lying around, but any lateral audio mosfets can be used.
The power requirements obviously mean this thing won't be portable but I'm hoping the performance will rival the Opamp based designs but with far more power.
Not included in the schem are the power supply + decoupling caps and diode protection on the mosfets.
As yet I have not begun the build but have run a few simulations and started to optimize the component values and the results look very promising
Output from this design is realistically only limited by your power supply; as shown the quiescent current in the mosfets is ~100mA which will give an entire watt of headroom with exceedingly low distortion. Max power for my phones is 200mW so I'll never run out of headroom (but I might blow them up) and the Iq could be lowered to 70 or 80mA. When too more current is drawn the amp will operate in class B up to the voltage limitation of the supply.
Output impedance is tiny.
Frequency response in sims is flat (+/-1dB) from less than 10Hz to in excess of 150kHz. THD according to Pspice is looking good but I am always skeptical of THD simulations, and I would be surprised if it turns out to be as low as the opamp headamps.
Gain is set at around 11.
The design, with few alterations, can be turned into a class A or AB power amp up to around 100WRms by changing only the Iq, using small sig. transistors capable of handling the voltage, a bigger power supply and sufficient heatsinks.
I know the power mosfets are huge overkill for this project, but I've been wanting something to do something with them for ages...
Anyway, If anyone has any feedback, observations or criticisms let me know!

Antoinel · 22nd February 2012, 02:01 PM

Quote:

Originally Posted by mnturner

Hey guys,
Max power for my phones is 200mW so I'll never run out of headroom (but I might blow them up) and the Iq could be lowered to 70 or 80mA. and sufficient heatsinks.

Go ahead and blow your headphones; but God forbid do not rupture your eardrums and loose your hearing. A deaf audiophile is unpleasant.

mnturner · 22nd February 2012, 03:25 PM

Can't disagree with that; a gain selector switch might help reduce the likelihood of an eardrum blasting.

sgrossklass · 22nd February 2012, 11:25 PM

Gain selection is, indeed, one of the most useful facilities on a headphone amp, assuming it's supposed to be reasonably universal. After all, sensitivity easily varies in a range of 40 dB.

I don't see why this design shouldn't be able to perform well. Keep in mind that you can afford luxurious amounts of VAS and input stage current.

As shown, the volume pot should be no more than 10k.

Since the gain in headphone amps can be quite low, the input stage might benefit from a cascode driven from the tail (in order to keep input transistor Vce constant and thus eliminate common-mode distortion), as suggested in chapter 4 of Mr. Self's handbook.

Antoinel · 23rd February 2012, 12:48 AM

A DIY project is inherently customized to the builder's needs and wants. Other DIYers may/will adopt this unique design and refine it further to suit their other requirements. At the outset, it is important to clearly define the following even before writing the schematic of the proposed headphone device:

What is the source of your music signal to the headphone device? Is it a compact disc player, a preamplifier, a tape, etc.
What is the output impedance of the music signal source? Is it low or close to zero Ohms or high e.g. 10K or higher?
Does the music signal source need a specific input impedance [or range] by your headphone device?
What is the range of the output voltage of the music signal source? Is it in the milliVolt or Volt range?
What is the impedance of your headphones? Is it as low as 8-16 Ohms or as high as 600 Ohms?
What is the maximum output voltage [or power in milliwatts] which will drive your headphones to [your] comfortable level of listening enjoyment?
Can you hear 10KHz or higher?
Will your headphone device protect your hearing and that of other DIYers should they adopt your design?

This ball is in your court!

mnturner · 23rd February 2012, 06:10 AM

Ok,
The amp is intended to accept a variety of sources, in my case they will mostly be computer sound cards, cd players and mp3 devices with low output impedance and about 2v pk-pk outputs.
The specced amp input impedance varies for these but if a line out with a specific requirement (as opposed to a headphone out designed to operate into 16 - 32 ohms) is used, input impedance of the amp can be lowere by addition of a resistor to ground before the input cap.
My particular phones are 64 ohm with 200mv max continuous rating; the amp can provide excess headroom for pretty much all dynamic headphone impedances.
Damage to headphones/ hearing is alway possible, but would be made less likely by the addition of a gain selector and awareness of the user.

mnturner · 23rd February 2012, 06:12 AM

Oh and no, I can only hear to about 15khz, and I'm not even 30 yet

mnturner · 23rd February 2012, 06:14 AM

Sgrossclass: you could definitely increase the vas current but given the low current needs of the mosfets I don't think it's necessary.

mnturner · 23rd February 2012, 10:28 AM

Also, that bootstrapped cascode setup Self suggests is interesting but he seems to suggest that the increase in HF distortion is present only at very low gain settings; he tested with 1-2 and at gain of 2 the effect was already reduced by a factor of about 5. At a gain of 11 I doubt that the effect would be significant.

Antoinel · 23rd February 2012, 11:43 AM

I am afraid your approach to this DYI ignores the teaching in the published article of "Gain Structure" by Michael Mardis. An input signal of 2 Vp-p is first [highly] attenuated and then amplified (by a factor of ?; could not tell from schematic) to generate an output signal of 0.1-2 Vp-p across your headphones. Of course, the overall design of your amp can readily do this. But; do you really need this potent amplifier [or any ] to do this proposed job? A more pertinent direction to this DIY is a power amp. At 63, I am still good to hear between [8 -10] KHz.

23rd February 2012, 12:48 AM	#5
Antoinel diyAudio Member Join Date: Jan 2012 Location: Pennsylvania Blog Entries: 1	A DIY project is inherently customized to the builder's needs and wants. Other DIYers may/will adopt this unique design and refine it further to suit their other requirements. At the outset, it is important to clearly define the following even before writing the schematic of the proposed headphone device: What is the source of your music signal to the headphone device? Is it a compact disc player, a preamplifier, a tape, etc. What is the output impedance of the music signal source? Is it low or close to zero Ohms or high e.g. 10K or higher? Does the music signal source need a specific input impedance [or range] by your headphone device? What is the range of the output voltage of the music signal source? Is it in the milliVolt or Volt range? What is the impedance of your headphones? Is it as low as 8-16 Ohms or as high as 600 Ohms? What is the maximum output voltage [or power in milliwatts] which will drive your headphones to [your] comfortable level of listening enjoyment? Can you hear 10KHz or higher? Will your headphone device protect your hearing and that of other DIYers should they adopt your design? This ball is in your court!

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
my Class-A MOSFET headphone amp	davygrvy	Headphone Systems	28	18th October 2011 06:42 AM
All Discrete Class-A Headphone Amp	mnturner	Headphone Systems	33	6th June 2011 01:39 PM
mos/bjt buffer in class A for headphone amp	sunrise	Headphone Systems	7	20th October 2008 06:34 PM
Lateral mosfet w Normal or Diamond buffer?	fab	Solid State	22	8th June 2007 04:23 PM
darlington/mosfet in emittor follower / headphone amp	svokke	Headphone Systems	11	8th June 2005 06:19 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

22nd February 2012, 03:25 PM	#3
mnturner diyAudio Member Join Date: May 2010	Can't disagree with that; a gain selector switch might help reduce the likelihood of an eardrum blasting.

22nd February 2012, 11:25 PM	#4
sgrossklass diyAudio Member Join Date: Nov 2004 Location: Germany	Gain selection is, indeed, one of the most useful facilities on a headphone amp, assuming it's supposed to be reasonably universal. After all, sensitivity easily varies in a range of 40 dB. I don't see why this design shouldn't be able to perform well. Keep in mind that you can afford luxurious amounts of VAS and input stage current. As shown, the volume pot should be no more than 10k. Since the gain in headphone amps can be quite low, the input stage might benefit from a cascode driven from the tail (in order to keep input transistor Vce constant and thus eliminate common-mode distortion), as suggested in chapter 4 of Mr. Self's handbook.

23rd February 2012, 06:10 AM	#6
mnturner diyAudio Member Join Date: May 2010	Ok, The amp is intended to accept a variety of sources, in my case they will mostly be computer sound cards, cd players and mp3 devices with low output impedance and about 2v pk-pk outputs. The specced amp input impedance varies for these but if a line out with a specific requirement (as opposed to a headphone out designed to operate into 16 - 32 ohms) is used, input impedance of the amp can be lowere by addition of a resistor to ground before the input cap. My particular phones are 64 ohm with 200mv max continuous rating; the amp can provide excess headroom for pretty much all dynamic headphone impedances. Damage to headphones/ hearing is alway possible, but would be made less likely by the addition of a gain selector and awareness of the user.

23rd February 2012, 06:12 AM	#7
mnturner diyAudio Member Join Date: May 2010	Oh and no, I can only hear to about 15khz, and I'm not even 30 yet

23rd February 2012, 06:14 AM	#8
mnturner diyAudio Member Join Date: May 2010	Sgrossclass: you could definitely increase the vas current but given the low current needs of the mosfets I don't think it's necessary.

23rd February 2012, 10:28 AM	#9
mnturner diyAudio Member Join Date: May 2010	Also, that bootstrapped cascode setup Self suggests is interesting but he seems to suggest that the increase in HF distortion is present only at very low gain settings; he tested with 1-2 and at gain of 2 the effect was already reduced by a factor of about 5. At a gain of 11 I doubt that the effect would be significant.

23rd February 2012, 11:43 AM	#10
Antoinel diyAudio Member Join Date: Jan 2012 Location: Pennsylvania Blog Entries: 1	I am afraid your approach to this DYI ignores the teaching in the published article of "Gain Structure" by Michael Mardis. An input signal of 2 Vp-p is first [highly] attenuated and then amplified (by a factor of ?; could not tell from schematic) to generate an output signal of 0.1-2 Vp-p across your headphones. Of course, the overall design of your amp can readily do this. But; do you really need this potent amplifier [or any ] to do this proposed job? A more pertinent direction to this DIY is a power amp. At 63, I am still good to hear between [8 -10] KHz.